This competing continuation proposal examines changes in suicidal behaviors and related risk and protective factors. Suicide is a major cause of death among youth aged 15-24 years. Suicide rates increase sharply from 20-24 years. Potential high school dropouts are known to be at higher risk than average for suicide and suicidal behaviors. Research indicates they enter young adulthood with greater risk and fewer protective factors. What is unknown is the impact of adolescent suicidal behaviors on young adult outcomes. This proposal describes a 4-wave, longitudinal study of high-risk youth across young adulthood (18-23 years). Study aims are to 1) test the temporal stability of a measurement model of suicide potential in adolescents vs. young adults; 2) study the longitudinal effects of adolescent risk and protective factors on the health outcomes of emotional distress, drug involvement, functional status, and suicidal behaviors; and to 3) examine patterns of change in suicide risk and related risk and protective factors across young adulthood, identifying predictors of these patterns of change. Subjects will be 1,150 young adults (18-23 years), identified as potential high school dropouts in the parent grant, and comprehensively assessed for suicide potential when 15-18 years old. The sample is approximately 48 percent female and 60 percent minority. Subjects will be surveyed annually 4 times and will participate in a comprehensive, suicide-risk assessment interview the final year. Young adult outcome variables include suicidal behaviors (thoughts/threats/attempts) and the related risk factors of emotional distress, drug involvement and functional status. Mediators include personal and social protective factors (coping/support). Three analytic approaches will be used: 1) confirmatory factor analyses to test factor stability of outcome construct; 2) structural equation modeling to test hypothesized relationships among adolescent antecedents and young adult outcomes; and 3) latent growth curve modeling (LGM) to determine patterns and predictors of change in suicidal behaviors across young adulthood.